2024-03-29T09:38:16Z
http://www.ijsom.com/?_action=export&rf=summon&issue=508
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
The Comparison of Neural Networks’ Structures for Forecasting
Ilham
Slimani
Ilhame
El Farissi
Said
Achchab
This paper considers the application of neural networks to demand forecasting in a simple supply chain composed of a single retailer and his supplier with a game theoretic approach. This work analyses the problem from the supplier’s point of view and the employed dataset in our experimentation is provided from a recognized supermarket in Morocco. Various attempts were made in order to optimize the total network error and the findings indicate that different neural net structures can be used to forecast demand such as Adaline, Multi-Layer Perceptron (MLP), or Radial Basis Function (RBF) Network. However, the most adequate one with optimal error is the MLP architecture.
Neural networks
Artificial intelligence
Supply Chain Management
information sharing
Demand forecasting
Game theory
2017
05
01
105
114
http://www.ijsom.com/article_2729_a658b3f1087d490d523d43b0d828838b.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
Design of Forward/reverse Logistics with Environmental Consideration
Masoud
Rabbani
Niloufar
Akbarian Saravi
Hamed
Farrokhi-Asl
Growth of environmental issues has caused to consider various factors that influence on condition of environment. Green supply chain has absorbed care of researchers because of its considerable impacts on environment. In this regard, this study designs the forward/revers logistics network by putting emphasis on environmental aspects in its model such as quantity of CO2 emission. In this logistics network, three objective functions such as minimizing the total cost and quantity of CO2 emission as well as maximizing the satisfaction of customers have been considered, simultaneously. Because of considering of three objective functions in this model, multi objective optimization methods persuade the researchers to implement them. Non-dominated sorting genetic algorithms (NSGA-ӀӀ) and Multi-objective particle swarm optimization (MOPSO) are proposed to cope with this problem. The results acquired from experiments on several test problems are verified by GAMS software. Finally, the results obtained through experiments on different problems verify the superiority of NSGA-ӀӀ over MOPSO in terms of all comparison metrics.
Green supply chain
CO2 emission
Forward/reverse logistics
Environmental issues
Multi objective optimization
2017
05
01
115
132
http://www.ijsom.com/article_2726_809642bf10c6ca5a9e26ab50b37141f1.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
Modeling the Multi-period and Multi-product Closed-loop Supply Chain Network Design Problem Considering Reused Cost and Capacity Constraints
Ali
Derakhshan
Seyyed Mohammad Hassan
Hosseini
Ali Akbar
Hassani
The importance of remanufacturing used products into new ones has been widely recognized in the literature and in practice. This is due to both of economic opportunities and environmental aspects. This paper aims to design a new integrated multi-period & multi-product closed-loop supply chain network considering reused cost and capacity constraints for all stages. In this problem the stages of supplier, assembler, retailer, customer, collection center, refurbishing center, and disassembler is regarded consequently. The considered objective function is total cost factors that consists of 7 components: costs of associated with locating the plants and retailers, purchasing, transportation, collection of used products from customers, disposal for subassemblies, refurbishing, and finally refund to customers. First, parameters and decision variables of this problem are defined, then a mixed integer linear programming mathematical model is presented. The proposed mathematical model is run applying the GAMS software. Two real examples (shed light, and power-outlet) are considered to solve using the proposed mathematical model. These two examples were obtained based on data in two new references. Since this problem is known as NP-Hard, the model is run just for small-sized problem consists of four suppliers, two disassemblers, two retailers, and two periods. The results are analysed and some sensitivity analysis have been done for the effective factors. These result show that, the demand has a less effect on total cost. But Purchasing/refurbishing cost ratio has a high effect on the objective function. Finally, the capacity of collection and refurbishing centers has a high effect in primary changes and this effect gradually reduced. So having the proper capacity for collection and refurbishing centers and also creating balance between different stages can reduce overall cost.
Capacity constraints
Closed-loop supply chains
Reused costs
2017
05
01
133
149
http://www.ijsom.com/article_2727_f57f3e8fc5188c47623df4a9c5c75304.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
On Solutions of Possibilistic Multi- objective Quadratic Programming Problems
Hamiden
Khalifa
In this paper, a multi- objective quadratic programming (Poss- MOQP) problem with possibilistic variables coefficients matrix in the objective functions is studied. Through the use of level sets the Poss- MOQP problem is converted into the corresponding deterministic multi- objective quadratic programming ( MOQP) problem and hence into the single parametric quadratic programming problem using the weighting method. An extended possibly efficient solution is specified. A necessary and sufficient condition for finding such a solution is established. A relationship between the solutions of possibilistic levels is constructed. Numerical example is given in the sake for the paper to clarify the obtained results.
Multi- objective quadratic programming
Possibilistic variables
Possibilistic efficient solution
level set
Possibly efficient solution
Possibly optimal solution
2017
05
01
150
157
http://www.ijsom.com/article_2728_61dedbc786adb8107bbd87fae493239c.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
Modeling and Solving a Blood Supply Chain Network: An approach for Collection of Blood
Hassan
Heidari-Fathian
Seyed Hamidreza
Pasandideh
Management of the blood as a vital and scarce resource is very important. The aim of this research is to present a novel mathematical model for designing a reliable blood supply chain network. This network consists of three main echelons including donors, collection facilities and demand points. At the collection echelon, three types of facilities are considered for receiving the bloods from the donors: main blood centers (MBCs), demountable collection centers (DCCs), and mobile blood facilities (MBFs). DCCs, and MBFs are mobile facilities that don’t have a permanent location and always move from a location to another one for collecting the bloods from the donors. The main difference between the MBFs and DCCs is that the DCCs can only visit at most a candidate location every period, but the MBFs can visit more than one candidate location in every period. Also, there is differences between their capacities and their costs. Both of DCCs and MBFs dispatch the collected bloods to the MBCs that are permanent facilities and are responsible for receiving the bloods and performing the blood transfusion process and finally sending the bloods to the demand points. Using a numerical example, the applicability of the proposed network is analyzed.
Blood supply chain
Perishable product
Mathematical programming
GAMS
Supply Chain Management
2017
05
01
158
166
http://www.ijsom.com/article_2725_c5da0639c7320c3305fdc915afa780f3.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
A New Mathematical Model for Simultaneous Lot-sizing and Production Scheduling Problems Considering Earliness/Tardiness Penalties and Setup Costs
Parinaz
Vaez
This paper investigated the problem of simultaneous determination of lot-sizing and production scheduling with earliness/tardiness penalties. In this problem, decisions about lot-sizing and scheduling are made so that the sum of holding, tardiness, and setup costs is minimized. There are n orders waiting to be processed on a machine. Each order has its own due date as well as tardiness and earliness cost being the same as holding cost .Each order is delivered only once. If the production is completed before or on the due date, delivery will be on the due date. Otherwise, the order will be delivered immediately after its production is completed. In spite of its wide applications, this problem has not yet been reported in the literature. A mathematical model was presented as solution methods for the problem. Two meta-heuristics, namely, Simulated Annealing and Ant Colony System meta-heuristic algorithms are presented for solving the problem. Also, lower bounds are obtained from solving the problem relaxation, and they are compared with the optimal solutions to estimate the goodness of two meta-heuristic algorithms. They are difficult benchmarks, widely used to measure the efficiency of metaheuristics with respect to both the quality of the solutions and the central. The results show that the Simulated Annealing recorded a lower solution time and average percentage deviation than did the Ant Colony System algorithm. The presented SA is capable to solve large instances that are mostly compatible with the real-world problems.
Scheduling
Lot-sizing
Earliness/Tardiness
Simulated Annealing
Ant Colony System
2017
05
01
167
179
http://www.ijsom.com/article_2731_6192678db270ee2c1f52721aca1ff6b1.pdf
International Journal of Supply and Operations Management
IJSOM
23831359
23831359
2017
4
2
Determining of an Optimal Maintenance Policy for Three State Machine Replacement Problem Using Dynamic Programming
Mohammad Saber
Fallahnezhad
Morteza
Pourgharibshahi
In this article, we present a sequential sampling plan for a three-state machine replacement problem using dynamic programming model. We consider an application of the Bayesian Inferences in a machine replacement problem. The machine was studied at different states of good, medium and bad. Discount dynamic programming (DDP) was applied to solve the three-state machine replacement problem, mainly to provide a policy for maintenance by considering item quality and to determine an optimal threshold policy for maintenance in the finite time horizon. A decision tree based on the sequential sampling which included the decisions of renew, repair and do-nothing was implemented in order to achieve a threshold for making an optimized decision minimizing expected final cost. According to condition-based maintenance, where the point of defective item is placed in continuing sampling area, we decided to repair the machine or to continue sampling. A sensitivity analysis technique shows that the optimal policy can be very sensitive.
Machine replacement
Dynamic programming
Sequential sampling plan
Maintenance
2017
05
01
180
192
http://www.ijsom.com/article_2730_0472c06ac5a83a39fbb39a71d569304d.pdf